Identification of women at risk for developing postmenopausal osteoporosis with vertebral fractures: role of history and single photon absorptiometry.

Putative risk factors for the development of postmenopausal osteoporosis (PMO) with vertebral fractures were examined in a retrospective study of 663 postmenopausal white females aged 45-75 years (266 women with non-traumatic vertebral compression fractures (VF+), 134 non-fractured women from a general medicine clinic (controls) and 263 non-fractured women who were evaluated when they presented specifically for osteoporosis screening (VF-]. The VF+ women differed from control women in several respects. The VF+ group reported a higher prevalence of a positive family history of osteoporosis, and a higher prevalence of a history of medical or surgical conditions known to be independently associated with metabolic bone disease, had fewer children, were smaller (weight, height) and were slightly older. The two groups, VF+ and controls, did not differ with respect to cigarette smoking, alcohol consumption, exercise habits, menstrual or menopausal history, dietary intake of milk and cheese or in amount taking calcium supplements during pregnancy. The VF+ group also differed in certain respects from the VF- group. The VF+ group were smaller (weight, height) and were older. The VF+ group had lower cortical bone mass (measured by single photon absorptiometry of the non-dominant forearm) than either the control or VF- groups. The latter two groups did not differ from each other with respect to this measurement. These markers demonstrated limited sensitivity and specificity as estimated from a confirmatory data set, particularly for the historical and anthropometric variables. We conclude that an assessment of the risk of developing PMO with vertebral fractures cannot be based on the putative risk factors as measured in our study, but must be based on measurement of bone mass.